csvsql 0.1.0 → 0.1.1
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/.gitignore +2 -0
- data/Gemfile.lock +1 -1
- data/README.md +3 -3
- data/exe/csvsql +18 -2
- data/lib/csvsql/db.rb +112 -109
- data/lib/csvsql/tracker.rb +44 -0
- data/lib/csvsql/version.rb +1 -1
- data/lib/csvsql.rb +7 -2
- metadata +3 -2
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 70fa4c1c358f92a85d5f1cd0502081297d1d27cc
|
4
|
+
data.tar.gz: 0bf5efcd968a8965c8308b75e39ae82e43c8d29e
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 88ddda2f38d83fd4defc23bbfd0d596517f600aacf30963a63b1a15c7f5dca73d7dfe977b0d8b80bff597238ee8e515f559480afb6d54c6a3c2b30e7d6626f24
|
7
|
+
data.tar.gz: d51eb995c1acfe9420153e3acb25f91c06137276f88dd3d67d60d7807395acefd9a7d9403030dfde5a22ab3c0e3492e5de44d479478eca26288f0b475e30b4e4
|
data/.gitignore
CHANGED
data/Gemfile.lock
CHANGED
data/README.md
CHANGED
@@ -75,10 +75,10 @@ csvsql -i mydata.csv "select name, total from csv where total < 30" | csvsql "se
|
|
75
75
|
It will save the CSV data to a tempfile. we use `~/.csvsql_cache` folder to save the cache
|
76
76
|
|
77
77
|
```
|
78
|
-
csvsql -i large.csv -
|
78
|
+
csvsql -i large.csv -c "select count(*) from csv"
|
79
79
|
|
80
80
|
# the second, it will be fast.
|
81
|
-
csvsql -i large.csv -
|
81
|
+
csvsql -i large.csv -c "select count(*) from csv"
|
82
82
|
```
|
83
83
|
|
84
84
|
### Clear Cache
|
@@ -86,7 +86,7 @@ csvsql -i large.csv -t "select count(*) from csv"
|
|
86
86
|
This command will remove all data in the `~/.csvsql_cache`
|
87
87
|
|
88
88
|
```
|
89
|
-
csvsql --clear
|
89
|
+
csvsql --clear-cache
|
90
90
|
```
|
91
91
|
|
92
92
|
|
data/exe/csvsql
CHANGED
@@ -8,7 +8,8 @@ require 'csvsql'
|
|
8
8
|
|
9
9
|
options = {}
|
10
10
|
OptionParser.new do |opts|
|
11
|
-
opts.banner = "
|
11
|
+
opts.banner = "Csvsql #{Csvsql::VERSION}\nUsage: csvsql [options] SQL"
|
12
|
+
opts.version = Csvsql::VERSION
|
12
13
|
|
13
14
|
opts.on('-i', '--input path', "CSV file path, optional. read from stdin if no give") do |path|
|
14
15
|
options[:csv_path] = path
|
@@ -18,9 +19,20 @@ OptionParser.new do |opts|
|
|
18
19
|
options[:use_cache] = true
|
19
20
|
end
|
20
21
|
|
22
|
+
opts.on(
|
23
|
+
'-b', '--batch-rows n',
|
24
|
+
"How many rows to import per batch. Default value is #{Csvsql::Db::BATCH_ROWS}"
|
25
|
+
) do |n|
|
26
|
+
options[:batch_rows] = n.to_i
|
27
|
+
end
|
28
|
+
|
21
29
|
opts.on('--clear-cache', "Clear all cache data") do
|
22
30
|
options[:clear_cache] = true
|
23
31
|
end
|
32
|
+
|
33
|
+
opts.on('--debug', "Print debug info") do
|
34
|
+
options[:debug] = true
|
35
|
+
end
|
24
36
|
end.parse!
|
25
37
|
|
26
38
|
if options[:clear_cache]
|
@@ -29,6 +41,10 @@ if options[:clear_cache]
|
|
29
41
|
exit
|
30
42
|
end
|
31
43
|
|
44
|
+
if options[:debug]
|
45
|
+
Csvsql::Tracker.tracker = Csvsql::Tracker.new(Logger.new($stdout))
|
46
|
+
end
|
47
|
+
|
32
48
|
csv_data = options[:csv_path] || StringIO.new($stdin.read)
|
33
49
|
|
34
|
-
puts Csvsql.execute(ARGV[0], csv_data, use_cache: options[:use_cache])
|
50
|
+
puts Csvsql.execute(ARGV[0], csv_data, use_cache: options[:use_cache], batch_rows: options[:batch_rows])
|
data/lib/csvsql/db.rb
CHANGED
@@ -2,146 +2,149 @@
|
|
2
2
|
|
3
3
|
require 'digest'
|
4
4
|
|
5
|
-
|
6
|
-
|
7
|
-
|
8
|
-
|
5
|
+
class Csvsql::Db
|
6
|
+
BATCH_ROWS = 10000
|
7
|
+
CACHE_DIR = File.join(Dir.home, '.csvsql_cache')
|
8
|
+
FileUtils.mkdir_p(CACHE_DIR) unless Dir.exists?(CACHE_DIR)
|
9
9
|
|
10
|
-
|
10
|
+
attr_reader :use_cache, :csv_path, :csv_io, :db, :batch_rows
|
11
11
|
|
12
|
-
|
12
|
+
def self.clear_cache!
|
13
|
+
require 'fileutils'
|
14
|
+
FileUtils.rm_f(Dir.glob(File.join(CACHE_DIR, '*')))
|
15
|
+
end
|
13
16
|
|
14
|
-
|
15
|
-
|
16
|
-
|
17
|
-
|
17
|
+
def initialize(use_cache: false, batch_rows: nil)
|
18
|
+
@db = nil
|
19
|
+
@csv_io = nil
|
20
|
+
@csv_path = nil
|
21
|
+
@use_cache = use_cache
|
22
|
+
@batch_rows = batch_rows || BATCH_ROWS
|
23
|
+
end
|
18
24
|
|
19
|
-
|
20
|
-
|
21
|
-
|
22
|
-
|
23
|
-
|
25
|
+
# action:
|
26
|
+
# raise: default
|
27
|
+
# exit
|
28
|
+
def sql_error_action=(action)
|
29
|
+
@sql_error_action = action.to_sym
|
30
|
+
end
|
24
31
|
|
25
|
-
|
26
|
-
|
27
|
-
|
28
|
-
|
29
|
-
|
30
|
-
end
|
32
|
+
def execute(sql)
|
33
|
+
db.execute(sql)
|
34
|
+
rescue SQLite3::SQLException => e
|
35
|
+
process_sql_error(sql, e)
|
36
|
+
end
|
31
37
|
|
32
|
-
|
33
|
-
|
34
|
-
|
35
|
-
|
36
|
-
|
38
|
+
def prepare(sql)
|
39
|
+
db.prepare(sql)
|
40
|
+
rescue SQLite3::SQLException => e
|
41
|
+
process_sql_error(sql, e)
|
42
|
+
end
|
37
43
|
|
38
|
-
|
39
|
-
|
40
|
-
|
41
|
-
|
44
|
+
def import(csv_data_or_path)
|
45
|
+
case csv_data_or_path
|
46
|
+
when StringIO, IO
|
47
|
+
@csv_io = csv_data_or_path
|
48
|
+
else
|
49
|
+
@csv_path = csv_data_or_path
|
42
50
|
end
|
51
|
+
@db = SQLite3::Database.new(get_db_path(@csv_path))
|
43
52
|
|
44
|
-
|
45
|
-
|
46
|
-
|
47
|
-
@csv_io = csv_data_or_path
|
48
|
-
else
|
49
|
-
@csv_path = csv_data_or_path
|
50
|
-
end
|
51
|
-
@db = SQLite3::Database.new(get_db_path(@csv_path))
|
52
|
-
|
53
|
-
tables = db.execute("SELECT name FROM sqlite_master WHERE type='table';").first
|
54
|
-
unless tables && tables.include?('csv')
|
55
|
-
init_db_by_csv(@csv_io ? CSV.new(@csv_io) : CSV.open(@csv_path))
|
56
|
-
end
|
57
|
-
true
|
53
|
+
tables = db.execute("SELECT name FROM sqlite_master WHERE type='table';").flatten
|
54
|
+
unless tables.include?('csv')
|
55
|
+
init_db_by_csv(@csv_io ? CSV.new(@csv_io) : CSV.open(@csv_path))
|
58
56
|
end
|
57
|
+
true
|
58
|
+
end
|
59
59
|
|
60
|
-
|
60
|
+
private
|
61
61
|
|
62
|
-
|
63
|
-
|
64
|
-
|
65
|
-
|
66
|
-
end
|
62
|
+
def parser_header(csv_header)
|
63
|
+
csv_header.map do |col, r|
|
64
|
+
name, type = col.strip.split(':')
|
65
|
+
[name, (type || 'varchar(255)').downcase.to_sym]
|
67
66
|
end
|
67
|
+
end
|
68
68
|
|
69
|
-
|
70
|
-
|
69
|
+
def init_db_by_csv(csv)
|
70
|
+
header = parser_header(csv.readline)
|
71
71
|
|
72
|
-
|
73
|
-
|
74
|
-
|
72
|
+
cols = header.map { |name, type| "#{name} #{type}" }.join(', ')
|
73
|
+
sql = "CREATE TABLE csv (#{cols});"
|
74
|
+
execute sql
|
75
75
|
|
76
|
-
|
77
|
-
|
78
|
-
|
79
|
-
|
80
|
-
|
81
|
-
|
82
|
-
|
83
|
-
|
84
|
-
|
76
|
+
cache = []
|
77
|
+
col_names = header.map(&:first)
|
78
|
+
Csvsql::Tracker.commit(:import_csv)
|
79
|
+
csv.each do |line|
|
80
|
+
cache << line.each_with_index.map { |v, i| format_sql_val(v, header[i][1]) }
|
81
|
+
|
82
|
+
if cache.length >= batch_rows then
|
83
|
+
import_lines(cache, col_names)
|
84
|
+
cache.clear
|
85
85
|
end
|
86
|
-
import_lines(cache, col_names) unless cache.empty?
|
87
|
-
db
|
88
86
|
end
|
87
|
+
import_lines(cache, col_names) unless cache.empty?
|
88
|
+
Csvsql::Tracker.commit(:import_csv)
|
89
|
+
db
|
90
|
+
end
|
89
91
|
|
90
|
-
|
91
|
-
|
92
|
-
|
93
|
-
|
92
|
+
def import_lines(lines, col_names)
|
93
|
+
sql = Csvsql::Tracker.commit(:generate_import_sql) do
|
94
|
+
s = "INSERT INTO csv (#{col_names.join(', ')}) VALUES "
|
95
|
+
s += lines.map { |line| "(#{line.join(',')})" }.join(', ')
|
94
96
|
end
|
97
|
+
Csvsql::Tracker.commit(:execute_import_sql) { execute sql }
|
98
|
+
end
|
95
99
|
|
96
|
-
|
97
|
-
|
98
|
-
|
99
|
-
|
100
|
-
|
101
|
-
|
102
|
-
|
103
|
-
|
104
|
-
end
|
105
|
-
rescue => e
|
106
|
-
process_sql_error("Parse val: #{val}", e)
|
100
|
+
def format_sql_val(val, type)
|
101
|
+
case type
|
102
|
+
when :int, :integer then val.to_i
|
103
|
+
when :float, :double then val.to_f
|
104
|
+
when :date then "'#{Date.parse(val).to_s}'"
|
105
|
+
when :datetime then "'#{Time.parse(val).strftime('%F %T')}'"
|
106
|
+
else
|
107
|
+
"'#{val.gsub("'", "''")}'"
|
107
108
|
end
|
109
|
+
rescue => e
|
110
|
+
process_sql_error("Parse val: #{val}", e)
|
111
|
+
end
|
108
112
|
|
109
|
-
|
110
|
-
|
113
|
+
def process_sql_error(sql, err)
|
114
|
+
$stderr.puts(sql)
|
111
115
|
|
112
|
-
|
113
|
-
|
114
|
-
|
115
|
-
|
116
|
-
|
117
|
-
end
|
116
|
+
if @error_action == :exit
|
117
|
+
$stderr.puts(e.message)
|
118
|
+
exit
|
119
|
+
else
|
120
|
+
raise err
|
118
121
|
end
|
122
|
+
end
|
119
123
|
|
120
|
-
|
121
|
-
|
122
|
-
|
123
|
-
|
124
|
-
|
125
|
-
|
126
|
-
|
127
|
-
|
128
|
-
|
129
|
-
|
130
|
-
|
131
|
-
|
132
|
-
|
133
|
-
|
134
|
-
else
|
135
|
-
FileUtils.rm(cache_path)
|
136
|
-
cache_path
|
137
|
-
end
|
124
|
+
def get_db_path(csv_path)
|
125
|
+
csv_path = csv_path || ''
|
126
|
+
return '' unless File.exist?(csv_path)
|
127
|
+
|
128
|
+
if use_cache
|
129
|
+
stat = File.stat(csv_path)
|
130
|
+
filename = Digest::SHA2.hexdigest(File.absolute_path(csv_path)) + '.cache'
|
131
|
+
file_stat = [File.absolute_path(csv_path), stat.size, stat.ctime].join("\n")
|
132
|
+
stat_path = File.join(CACHE_DIR, filename.gsub(/\.cache$/, '.stat'))
|
133
|
+
cache_path = File.join(CACHE_DIR, filename)
|
134
|
+
|
135
|
+
if File.exist?(stat_path)
|
136
|
+
if File.read(stat_path) == file_stat
|
137
|
+
cache_path
|
138
138
|
else
|
139
|
-
|
139
|
+
FileUtils.rm(cache_path)
|
140
140
|
cache_path
|
141
141
|
end
|
142
142
|
else
|
143
|
-
|
143
|
+
File.write(stat_path, file_stat)
|
144
|
+
cache_path
|
144
145
|
end
|
146
|
+
else
|
147
|
+
''
|
145
148
|
end
|
146
149
|
end
|
147
150
|
end
|
@@ -0,0 +1,44 @@
|
|
1
|
+
require 'logger'
|
2
|
+
|
3
|
+
class Csvsql::Tracker
|
4
|
+
attr_reader :stats, :logger
|
5
|
+
|
6
|
+
def self.tracker
|
7
|
+
@tracker ||= new
|
8
|
+
end
|
9
|
+
|
10
|
+
def self.tracker=(t)
|
11
|
+
@tracker = t
|
12
|
+
end
|
13
|
+
|
14
|
+
def self.commit(*args, &block)
|
15
|
+
tracker.commit(*args, &block)
|
16
|
+
end
|
17
|
+
|
18
|
+
def initialize(logger = Logger.new('/dev/null'))
|
19
|
+
@stats = {}
|
20
|
+
@logger = logger
|
21
|
+
end
|
22
|
+
|
23
|
+
def commit(id, output: true, &block)
|
24
|
+
id = id.to_s
|
25
|
+
old = stats[id]
|
26
|
+
stats[id] = get_stat
|
27
|
+
|
28
|
+
if block
|
29
|
+
block.call.tap { commit(id) }
|
30
|
+
elsif output && old
|
31
|
+
logger.info("[#{id}] #{compare_stat(old, stats[id])}")
|
32
|
+
end
|
33
|
+
end
|
34
|
+
|
35
|
+
private
|
36
|
+
|
37
|
+
def get_stat
|
38
|
+
{ time: Time.now }
|
39
|
+
end
|
40
|
+
|
41
|
+
def compare_stat(old, new)
|
42
|
+
"Time cost: #{((new[:time] - old[:time]) * 1000000).to_i / 1000}ms"
|
43
|
+
end
|
44
|
+
end
|
data/lib/csvsql/version.rb
CHANGED
data/lib/csvsql.rb
CHANGED
@@ -6,15 +6,20 @@ require 'csv'
|
|
6
6
|
require 'sqlite3'
|
7
7
|
|
8
8
|
require 'csvsql/db'
|
9
|
+
require 'csvsql/tracker'
|
9
10
|
|
10
11
|
module Csvsql
|
11
12
|
def self.execute(sql, csv_data, opts = {})
|
12
13
|
csvdb = Csvsql::Db.new(opts)
|
13
14
|
csvdb.import(csv_data)
|
14
|
-
pst =
|
15
|
+
pst = Csvsql::Tracker.commit(:execute_query_sql) do
|
16
|
+
csvdb.prepare(sql)
|
17
|
+
end
|
18
|
+
Csvsql::Tracker.commit(:output_format)
|
15
19
|
CSV.generate do |csv|
|
16
20
|
csv << pst.columns.zip(pst.types).map { |c| c.compact.join(':') }
|
17
21
|
pst.each { |line| csv << line }
|
18
|
-
end
|
22
|
+
end.tap { Csvsql::Tracker.commit(:output_format) }
|
19
23
|
end
|
20
24
|
end
|
25
|
+
|
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: csvsql
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.1.
|
4
|
+
version: 0.1.1
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- jiangzhi.xie
|
8
8
|
autorequire:
|
9
9
|
bindir: exe
|
10
10
|
cert_chain: []
|
11
|
-
date: 2018-07-
|
11
|
+
date: 2018-07-19 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: sqlite3
|
@@ -102,6 +102,7 @@ files:
|
|
102
102
|
- exe/csvsql
|
103
103
|
- lib/csvsql.rb
|
104
104
|
- lib/csvsql/db.rb
|
105
|
+
- lib/csvsql/tracker.rb
|
105
106
|
- lib/csvsql/version.rb
|
106
107
|
homepage: https://github.com/xiejiangzhi/csvsql
|
107
108
|
licenses:
|